Agent Lightning v1.0

Agent Lightning v1.0.1 launches the Agent Lightning Skill, enabling coding agents to refine other AI agents by taking an editable agent and a benchmark and iteratively improving prompts, tools, workflows, models, and reasoning settings to balance accuracy, cost, latency, and reliability. The release adds CI, packaging, release automation, documentation, and benchmark reporting, and can be installed for Claude Code, Codex, or GitHub Copilot via a single command.